US1794246A - Electrical apparatus - Google Patents

Description

Feb. 24, 1931. J. c. READ 3 3 ELECTRICAL APPARATUS Filed Dec. a1. 1927 Fig. l.

Inventor: John C. Read His Attorney Patented Feb. 24, 1931 UNITED STATES PATENT OFFICE JOHN C. READ, OF CREWKERNE, ENGLAND, ASSIGNOR '10 GENERAL ELECTRIC COM- PANY, A CORPORATION OF NEW YORK ELECTRICAL APPARATUS Application filed December 31, 1927, Serial No. 244,004, and in Great Britain January 21, 1927.

My invention relates to electrical apparatus comprisin a rectifier and a plurality of main trans ormer windings intercon-v nected through the windings of an interphase transformer, and has for its principal object the provision of an improved arrange ment of the various transformer windings whereby the constructional and operating costs of such apparatus are greatly reduced.

In the construction of large rectifiers of the mercury vapor type, for example, it is sometimes desirable to supply the anodes of the rectifier with 12 or more phases, the multiplication of the number of phases having to be effected by the transformer.

The transformer secondary winding must necessarily possess a star point, since this star point forms the positive pole of the direct current system supplied by the rectifier. Asimple 12-phase star-connected secondary winding would have the disadvantage for this service that, owing to the small degree of overlap of the currents of successive anodes, and the high reactance between adjacent anodes, the transformer would be costly due to inefiicient utilization of its secondary winding, and the rate of fall of the direct current voltage with increasing load would be too rapid. It is possible to overcome this difficulty by splitting up the 12-phase secondary winding into four symmetrical 3- I phase star-connected systems, the star points of two of the systems being connected by. an inter-linking transformer, the star pointsof the other two systemsby another interlinking transformer, and the resulting star points of the two pairs of systems being connected by a third intcrlinkin g transformer.

This arrangement satisfactorily obtains the desired qualities of good utilization of the secondary winding and small voltage drop, but it has the disadvantage of requiring three separate interlinking transformers.

In accordance with the present invention this disadvantage is overcome by splitting up the 12 phase secondary winding into four symmetrical 3 phase star connected systems and bringing the leads from the respective neutral points each to one oft-he limbs of a four limbed transformer thereby obviating the expense of a plurality of transformers.

The economies effected by this arrangement are due to a number of causes. In the first place, the economy of material is greater because the same iron is used to carry the 180 cycle fluxes and the 360 cycle flux. The combination of these fluxes of different frequencies gives a total flux which is less than their arithmetical sum. At the same time, the combination of the two 180 cycle transformers in one unit results in the normal saving obtained by combining two single phase transformers into one polyphase trans former.

There as also a saving due to the fact that no coils have to be wound for currents larger than that coming from one of the neutrals of the main transformer, whereas the 360 cycle transformer must otherwise be wound for double this current, and, where large currents are involved, a coil of such a type is very expensive due to its heavy copper windlngs.

Another and very important economy results from the use of a single interphase transformer core structure of simple form in the place of three separate cores which must first be clamped independently and then fastened together so as to be placed in a common tank. By making the depth of the four-legged core somewhat greater than its width, it can be made to fit the clamps and tank of a standard three-phase transformer, thus allowing the use of parts which are standard for other types of transformers.

The invention will. be better understood from the following description when cousidered. in connection with the accompanying drawings and its scope will be pointed out in the appended claims.

In the accompanying drawings Figs. 1., and 7 diagrammatically illustrate different methods of carrying this invention into effect; and Figs. 3 to 6 and '8 relate to the operation of the apparatus of Figs. 1, 2 and 7.

In carrying out the invention as shown diagrammatically in Fig. 1, A, B, C and D are four 3-phase star-connected systems, each symmetrical as far as itself is concerned, and

"vector-1,- Fig. 3, represent the thir voltage -to be abs each displaced by 30 electrical degrees from any of the others, so that together they form a 12-phase secondary Winding. The leads from their respective neutral points are brought each to one leg of a -legged interphase transformer E, the windings on all the four legs of'E'being in the same direction and oined together at a common star point after Jassin throu h E. This common star. )Olllt transforn'ier on the alternating current components of the currentsflowing through it.

More generally, it is possible 'to obtain a similar effect in any polyphas'e star-connected system, by splitting up the system intoa number of separate star-connected systems, each symmetrical in itself,-"and connectingthe star points of these to an interphase transformer having a number of legs equal tOltllB, number of systems int-dwhichthe original polyphase systeniis split up, and connectin the leads from these separate star points to a common star point through this interphase' transformer, which is'connected either snnply in star or in-Zigzag-star. 1

The method of operation "of interphase transformers such asE and F is as follows: The function of the interphase transformer is to enable each of thestar-connected groups, such as A, to operate like a'three-phase rectifier. In orderto do this,the'windin'g of the interphase transformer has to absorb the difference voltage between therectified voltage produced by a three-phase group, such as A, and the arithmetical mean of the rectifiedvoltages produced by all the 'four groups A, B, C and D. This difference voltageconsists of a series of harmonic voltages, the orders of the harmonics including all the multiples of threeexeeptmulti les of twelve, the order of aharmonic beiii'g 'the ratio of its frequency 'to'the frequency of the supply; so that the harmonic whose order is a frequency three times the 's'iipply'fre'quency. The test of whether an interpha-sje transform-er such as E'or F is capable'of producingthe' desired-resuit is, therefore, to determine'whether is capable of'abso'rbing' all these harmonic volt- Considerin first ii the th rd'fha-rinoni'c, let

rbed' betwee- Lhannionic he 'neutral Since group B is displaced from group A, its third harmonic voltage Will be displaced 90,.from that produced by group A, and isreprese'nted by vector 2. Similarly, vectors 3 and 4 represent the third harmonic voltagesproduced bygroup's' C and D. Consequently, the four third harmonic voltages to be absorbed form a. symmetrical four- .phase system.v Since .E. is a four-legged "transformer,theflux in it will generate electromotive forces in its windings which will ;u-'holly--absorb the third harmonic voltages. In a similar way, 5, 6, 7 and 8, Fig. 4, are respectively the voltage-vectors ofth'e sixth harmonic voltages between A, B, Cand D and the main star point, and these a'lso are absorbed by theinterphase transformer; "Similarly, the ninth, fifteenth, eighteenth, and so on, harmonics are-absorbed by the interphase transformer. "For the twelfth,"twenty fourth, *and so on, harmonics, "however," t he vector diagram is as shown in Fig. 5, allthe four harmonic voltages tending tomagnetize the core ofthe interphase transformer in the same direction at every instant. Since the return path ifor this flux is through the air from yoke to yoke, itis clear that the interphase transformer will: not absorb 'these voltages 'but'this is of no importance since the twelfth, twenty fourtlnand so on,harmonics are not *amon'g those which "haveto'be ab 'sorbed. Consequently, the transformer'E absorbs all the harmonic voltages required'of it.

When the transformer F, Fig.2, is used instead of Qthetrans'former "E, for I the "third harmonic, the vector "diagram 'Gf'the magnetomotive forces produces 'on each 1eg=is .as "shown in Figt'6. 'Since'eaehleg of'the trans- "for-merFhas two windings producing magnetomotive forces 90 electrical degrees apart, the resulting magnetom'otive force on each leg a'gain forms a symmetrical four-phase system, and consequently thethird harmonic voltage is again successfully absorbed by the interphase transformer. Similar reasoning shows th at the transformerF' will also absorb all-the remaining harmonic voltages. present,

- namely, the sixth, nintln fiftee'nth an'd'so on.

fAdditional unwound legsmay beadded to any transformer with zigzag connection; such as the-transformer F, Without essentially "efthe number ofphases, p=m.n, where m and are integers -The. p -phase star-connected secondary wind ng is-then divided'into m mmet'rlc'al n phas'e "star-connected groups "displaced point 613 group A and the inainstari poiint.

points are connected together by an interphase tl'illlSfOIIIlQl havingm legs. Each leg may have one winding on it, connected to the neutral point of one of the m groups; or, each leg may have two windings in opposite directions on it, connected respectively in series with different ones of the m-star points, thus constituting a zigzag connection of the interphase transformer.

In choosing the type of zigzag connection, that is to say, in choosing which legs of the interphase transformer the two windings from a given one of the m groups shall be wound on, a number of alternative arrangements which will be satisfactory are always available, but certain arrangements have to be avoided. Thus, if the four-legged interphase transformer F, Fig. 2, had been conuectedas shown at F in Fig. 7, then the vector diagram of the magnetomotive forces produced on each leg by the sixth harmonic would have been as shown in Fig. 8, that is, no resulting magnetomotive force would be produced on any of the legs for this harmonic, and consequently the sixth harmonic voltage would not be absorbed. By this method of test, these zigzag connections which have to be avoided can easily be determined.

What I claim as new and desire to secure by Letters Patent of the United States, is:

1. A rectifying apparatus comprising four three phase star-connected systems each provided with a neutral terminal, a core having four legs, and a plurality of windings each wound on a different leg of said core and each connected between a common pointand the neutral terminal of a different one of said systems, to absorb the third harmonic of the supply frequency and all multiples of said harmonic except the twelfth and multiples thereof.

2. A system of connections for electric arc rectifiers having twelve or more phases, in which the p-phase star-connected secondary winding is divided into four symmetrical nphase star-connected groups displaced 360 7 electrical degrees apart, characterized by the feature that the star points of the four groups are connected together through a common interphase transformer having four legs, with one Winding on each leg, all the windings in the same direction, and each of the windings connected to a different one of the four star points, to absorb the nth harmonic of the supply frequency and all multiples of said harmonic except the 4 nth and multiples thereof.

3. A system of connections for electric arc rectifiers having twelve or more phases, in which the p-phase star-connected secondary winding is dividedinto four symmetrical n-phase star-connected groups dlsplaced P iv

Images